The Stylisation of Internet Life?: Predictors of Internet Leisure Patterns Using Digital Inequality and Status Group Perspectives

by Roderick Graham
City University of New York

Sociological Research Online 13(5)5
<http://www.socresonline.org.uk/13/5/5.html>
doi:10.5153/sro.1804

Received: 6 Aug 2008     Accepted: 4 Sep 2008    Published: 30 Sep 2008

Abstract

This research addresses the question: What are the predictors of internet leisure patterns? With the barriers to internet access receding the question can be asked: Whether or not social groups are beginning to distinguish themselves through different types of internet activities? This research will focus on the domain of internet leisure and entertainment within the United States population. Internet leisure is measured in this study by playing games, doing hobbies, reading, watching videos and listening to music. Predictors are drawn from a digital inequality and a Weberian status group perspective. Binary logistic regression models are run on a nationally representative survey from the Pew Research Center's Internet and American Life Project (N = 2013). This research suggests that: (1) both digital inequality and status group perspectives tend to work together to explain all internet activities, but a status group perspective provides more explanation for leisure patterns, (2) internet leisure is best conceptualised as a form of popular culture with minorities and people of low socioeconomic status consuming leisure at higher rates.

Keywords: Status Group, Digital Inequality, Digital Divide, Internet, Cultural Consumption, Culture, Leisure

Introduction

1.1 Research on digital inequality has shown consistently that the amount of time using the Internet, the quality of that usage and the benefits gained are not evenly distributed throughout the United States population (Attewell 2001; Warschauer 2003; Keegan 2004; Mossberger et al. 2006). People with lower incomes, who are less educated and are members of minority groups lag behind the rest of the population in internet usage. However, barriers to the Internet continue to recede and internet access is becoming the rule rather than the exception in even the most economically disadvantaged of United States households (National Telecommunication and Information Administration 2002; Martin and Robinson 2007). Also, a concerted effort has been made by the nation's leaders to decrease the gap between European-American and minority students, with some success (National Telecommunication and Information Administration 2002). With the barriers to internet access receding, the question can be asked whether or not status groups are beginning to distinguish themselves through different types of internet leisure activities.

1.2 Bourdieu argued in Distinction that when economic constraints are lifted, people's choices are more ordered by their own personal dispositions: 'As the objective distance from necessity grows, lifestyle increasingly becomes the product of what Weber calls a "stylization of life," a systematic commitment that orients and organizes the most diverse practices – the choice of vintage or a cheese or the decoration of a holiday home in the country.' (Bourdieu 1984 pp. 55 – 56)

1.3 This stylisation of life is not idiosyncratic, but ordered by a person's position in the class hierarchy—such that people who share class positions stylise their life similarly. This research will look for this stylisation of life by focusing on the domain of internet leisure and entertainment. Specifically, this research will address the question: 'What are the predictors of internet leisure patterns?'

1.4 I have selected two perspectives for guidance. The first is the research literature devoted to digital inequality. The second perspective is the Weberian notion that cultural patterns are associated with the unique preferences of status groups. Both perspectives imply that internet patterns mirror aspects of the social structure. The former implies that internet patterns reflect disparities in income and education, as well as differences between racial and ethnic groups. The latter implies that internet patterns are more reflective of lifestyle and social circumstances such as urban residency, gender, and marital status. These two perspectives will provide the predictors upon which to test differences in internet leisure patterns. I explain these perspectives in more detail below.

Literature Review

Digital Inequality and Internet Usage

2.1 As Servon and Nelson (2001) argue 'access to information technology (IT) and the ability to use it [has] increasingly become part of the toolkit necessary to participate and prosper in an information-based society' (279). The digital inequality perspective argues that people of higher income and higher education use the internet with greater frequency and for greater benefits (educational growth, greater employment opportunities) than middle and working class people. This argument has progressed from the initial Digital Divide debates (Attewell 2001; Norris 2001; Warschauer 2003; Kvavsny 2006) to the current argument about digital inequality. The former was concerned with raw hardware acquisition and the latter is concerned with the quality of usage (Dimaggio et al. 2004; Benkler 2006). Digital inequality assumes that income provides privileged access and education provides privileged understanding. Inequality is reproduced when members of the upper classes take disproportionate advantage of the benefits that the internet can offer.

2.2 These advantages can include distance learning, e-commerce, and access to privileged flows of knowledge through social networking. In a qualitative study of two neighbourhoods in the UK, Crang et al. (2006) find evidence for the development of a 'multispeed urbanism', where higher educated people were using communication technology as means to keep their fast-paced life cohesive. On the other hand, lower educated people used communication technology sporadically and for very specific one-time goals. The study by Crang et al. is emblematic of the majority of research that addresses the utilitarian uses of the Internet. Higher income, higher educated people are more high end users who manipulate the internet for greater benefits (National Telecommunications and Information Administration 2000, 2002; Hargittai 2008).

2.3 Although income and education are the major structuring principles in the digital inequality argument, racial differences are inextricably intertwined. For example, Chakraborty and Bosman (2005) measure PC ownership inequality within African-American and European-American households. Their study found that 'The poorest 50 per cent of white households indicate a PC ownership rate of approximately 40 per cent, while the poorest 50 per cent of African-American households reveal an ownership rate of about 27 per cent' (401). Gates (2000) has called this phenomenon the 'cyber-segregation' of minorities. Thus, racial differences are a major component of the body of digital inequality research (Servon and Nelson 2001; Warschauer 2003; Chakraborty and Bosman 2005; Mossberger et al. 2006; Ono and Zavodny 2008). Since the late 1990's, several initiatives were undertaken to redress differences in internet access (Lacey 2000; Kvasny 2006) between minorities (particularly African-Americans and Hispanic-Americans) and European-Americans. The gap between African-Americans and European-Americans is disappearing, especially in places where extreme poverty is not a factor (Mossberger et al. 2006). However, the gap between Hispanic speaking immigrants and European-Americans may be growing (Ono and Zavodny 2008). Because of the linkages between race/ethnicity, income, and education, all three of these predictors are included within the digital inequality perspective.

Status Groups and Internet Usage

2.4 The status group perspective provides an alternative to the digital inequality perspective[1] and is derived from the Weberian idea that people who are members of a status group tend to share a common lifestyle: '…status groups [italics in original] are normally communities…In contrast to the purely economically determined 'social situation' we wish to designate as 'status situation' every typical component of the life fate of men that is determined by a specific, positive or negative, social estimation of honor [italics in original]. This honor may be connected with any quality shared by a plurality…' (Weber in Gerth and Mills 1958: 186 – 187)

2.5 Importing Weber's term to the current research, it is being argued presently that as barriers to internet access continue to recede, patterns of internet usage are no longer a matter of affordability. Indeed, in other areas of research where the emphasis is on understanding cultural consumption patterns, status groups are more accurate predictors than economic related variables (van Eijck 2001; Chan and Goldthorpe 2007).

2.6 Status group differences have already been documented in recent research on internet usage. The main fault lines are gender and age. With respect to gender, Willoughby (2008) found that boys in high school were significantly more likely to use the internet than their female counterparts. Further, Spotts et al (1997) report lower use of instructional technology for women than men within a higher education faculty. In computer-mediated communication technologies such as chat rooms and e-mails, the types of written content are structured by gender (Gefen and Straub 1997; Koch et al. 2005). Ono and Zavodny (2003) reported that men use the Internet more intensely than women, although the gap between users and on-users has disappeared. Jackson et al (2001) assert that generally women communicate online and men search online. With respect to age, younger people are more frequent internet users than older people (Bimber 2000; Lorence and Park 2006; Xie and Jaeger 2008). While this age disparity is consistent, there are occasions when older people actively use the Internet. For example, Morrell et al. (2000) find that middle aged people use the Internet frequently for accessing health care information.

2.7 One particularly vibrant area of internet research is devoted to the implications of social networking. People who use the Internet for communication with friends and family can form 'virtual cliques', and exchange information and resources to the exclusion of others. This particular aspect of the Internet has been discussed widely (Rheingold 1993; Diani 2000; Brunting and Postmes 2002). These virtual cliques can extend into face-to-face interactions when people attend events planned online with other social networkers (Castells 2001). In general, the Internet has been seen as a means for social inclusion (Warschauer 2003), and does not lead to the conclusion that the Internet is creating 'haves' and 'have-nots' as much as it is creating 'likes' and 'dislikes'. For example, individuals who share the same sexual perspective (Garofalo et al. 2007), similar interests in social movements (Diani 2000; Brunsting and Postmes 2002; Adams and Roscgino 2005), or similar psychological dispositions (Morahan-Martin and Schumacher 2003) may network with each other.

Internet Leisure

2.8 I have shown above that the research devoted to IT and specifically the Internet can be divided into digital inequality and status group perspectives. However, the literature above does not address leisure or entertainment specifically. This reflects a general lack of emphasis on the entertainment and leisure dimensions of the Internet.

2.9However, a study done by Whitty and Mclaughlin (2007) found that people who were lonely or were expert users of the Internet were more likely to indulge in internet entertainment. Their study delineated three areas of entertainment: (1) pure computer entertainment, (2) facilitating future offline entertainment, and (3) for retrieving information about entertainment.

2.10 Because of the size and population of the survey sample (150 university students) more work needs to assess the reliability of these areas of leisure. Still a major insight of this work is that internet patterns, including those specifically related to entertainment, are sufficiently complex and cannot be understood as general internet usage or general internet frequency.

2.11 A burgeoning body of research is delving into internet gaming, where being male significantly predicts hours spent gaming (Schumacher and Morahan-Martin 2001; Durndell and Haag 2002; Willoughby 2008). Also, specific work on television viewing on the Internet (Deery 2003) suggests a popular, mass culture function for internet leisure. By citing examples of how game and news shows attempt to foster participation through internet surveys, Deery (2003) foresees the YouTube phenomenon when she suggests that the Internet is a reflection of the broader cultural trend away from elitist/modernist values and towards more populist and participatory values (162).

Method

Data

3.1 The analyses presented here are based on a nationally representative survey from the Pew Research Center's Internet and American Life Project (N = 2013). The survey, called 'Minor Moments' was conducted in November 2003. Data was collected through phone interviews. This survey asked questions about internet activities, beliefs about piracy, religious activity and socio-demographic characteristics. The socio-demographic and internet activity modules are the focus of the present research.

Independent Variables

3.2 The independent variables used in this analysis are derived from the two perspectives discussed above. I have also included a variable conceptualised as 'access' to control for the amount of time spent on the Internet. Research has shown that access is an important factor in internet usage (Katz and Rice 2002; Hargittai 2004). The access variable is home internet frequency.

3.3 Further, due to the characteristics of the Internet, leisure activities can be intertwined with other activities done online almost seamlessly. For example, someone may, in the course of sending an e-mail or purchasing products online, watch a video on a site such as YouTube, and then return to their e-mail draft or continue browsing merchandise. This is not completely dissimilar from other forms of leisure such as reading and watching television, where people routinely incorporate these items into their daily practices.[2]However, it does raise questions as to how people perceive their internet activities and whether or not it is easy to demarcate one activity from another.

3.4 In an effort to address this concern, variables that assess other types of internet activities will also be incorporated into the model and act as controls. These usage variables are 'gather information online', 'interact/communicate online', and 'do tasks/transactions online'. These variables will also be used as dependent variables in separate models and will be discussed in more detail in the subsequent section.

3.5 After controlling for access and other internet activities, the effects of digital inequality and status group variables can be assessed. The digital inequality variables are race/ethnicity[3], income, and education.[4] Status group variables are age, gender, community type and relationship status.

Dependent Variables

3.6 Two sets of dependent variables will be used to address the research question: 'What are the predictors of internet leisure patterns?' First, four general usage variables (information, communication, completing tasks and leisure) are treated as dependent variables. The first three of these variables are used as controls when predicting specific internet leisure patterns. However these usage activities are also treated as dependent variables in separate analyses. These four variables are binary, such that a respondent either reports doing this general activity, coded '1' or not, coded '0'.

3.7 Second, respondents are posed a set of question about five specific internet leisure activities. These activities are: playing games, doing hobbies, reading, watching videos and listening to music. The response categories were: 'I do this online', 'I do this both online and offline', 'I do this offline', and 'I don't do this at all'. Because I am interested in people who already do the activity, and 'stylise their consumption patterns by incorporating the Internet, individuals who reported not doing the activity at all were omitted. Further, I am interested in whether individuals choose to do this online or not. Thus, the two response categories 'I do this online' and 'I do this online and offline' are combined. The final dependent variables are five dummy variables coded '1' for those who do the activity and '0' for those who do not do the activity. See Table 1 for univariate statistics of independent and dependent variables.

Table 1. Univariate Statistics for Independent and Independent Variables (non-weighted)

Plan of Analysis

3.8 Binary logistic regression is used for all models estimated. Binary logistic regression is appropriate when the variables to be explained have two values. All dependent variables are binary in their outcomes—either a respondent reports liking or doing the activity (1) or not (0). The estimates presented are odds ratios, such that a unit change in the predictor (or for dummy variables such as race/ethnicity the presence of the predictor) is associated with a change in the odds the respondent will report doing the activity. Odds ratios over 1 indicate an increase of the odds, and estimates below 1 indicate a decrease in the odds. In the case of categorical variables such as race/ethnicity, an estimate over 1 indicates odds greater than that of the reference category and an estimate less than 1 indicates odds lesser than that of the reference category.

3.9 First, the general usage variables (information, communication, completing tasks and leisure) are estimated. This phase of the analysis is to assess the degree of difference (or similarity) between internet leisure activities and other types of activities. Rates of activity and significant predictors of the activity can be compared. Second, the five specific internet leisure activities (playing games, doing hobbies, reading, watching videos and listening to music) are estimated. All models use weights.[5] The formal interpretations of odds-ratios are cumbersome and do not reveal as much as simply comparing the magnitudes of the effects of predictors within models. Thus, when discussing regression models, expressing findings in explicit odds-ratios will be minimal.

Findings

Bivariate Statistics for General Internet Activity

4.1 In general, higher educated people, other Americans and younger respondents tend to do more of all activities. As a whole, people report doing information seeking and communicating more than completing tasks and seeking leisure. The groups that report doing the highest of any activity are college graduates (information seeking – 81 per cent, communication – 77 per cent) and people between the ages of 18 and 29 (information seeking – 79 per cent). One of the purposes of comparing all four types of internet activity is to look for differences with respect to leisure. Two notable differences between leisure and other categories is the drop-off in activity for college educated people and the overall low reports of activity in general.

Table 2. Percentages of Activity for General Internet Activity Variables (% are for those reporting doing the activity online)

Bivariate Statistics for Specific Leisure Activities

4.2People report higher percentages of game playing and doing hobbies than other leisure activities. Watching videos is the least reported activity (16 per cent of the population). A common pattern within specific leisure activities is that, generally, most people do not engage in internet leisure activities. No activity is done by a majority of the respondents. Further, only one activity—games, has any one group that has a majority of its respondents doing the activity. The college educated group also reports lower rates of specific activity than they do general activities. It appears as if college educated respondents report higher rates of general leisure activity, but when asked about specific activities their rates decrease.[6]

Table 3. Percentages of Activity for Specific Leisure Activities (% are for those reporting doing the activity online)

General Internet Activity

4.3 Several points are readily apparent from looking at the models in Table 4. I begin by discussing effects across all four models. I then discuss the differences between leisure and the other three activities modeled.

Table 4. Binary Logistic Regressions for General Internet Activities

4.4 First, as expected, an increase in access is associated with an increase in the probability of a respondent doing each activity. The odds of using the Internet for the general activities increase by factors ranging from 1.351 to 1.419 for one unit increase in home internet frequency.

4.5Unit increases in income increase the odds for the information and task activities, a finding that falls in line with the notion that people in higher socioeconomic strata are more high end users who manipulate the Internet for greater benefits (National Telecommunications and Information Administration 2000, National Telecommunications and Information Administration 2002; Hargittai 2008). Consistently we see that the odds of doing an activity are greater for highly educated people than for people with high school diplomas, with leisure being the only exception. Further, the magnitude of the effects of education is strong. The highest increase in odds for any independent variable on a dependent variable is the effect of higher education for the communication activity.

4.6 The odds for African-Americans are generally lower than for European – Americans (the reference group) and the odds for older groups are generally lower than for people between the ages of 18 and 29 (the reference group). These findings are consistent across all four activities. Status group effects other than age (community status, relationship status, and gender) are not significant in three of the four activities.

4.7 Leisure activity follows a different pattern than the other activities in two ways. First, education works in the opposite direction. College graduates are significantly different than high school graduates. For example, the odds of a college graduate is .687 the odds of a person with a high school diploma person doing leisure activities on the Internet. Second, status group variables become significantly associated, and the magnitude of status group effects are higher than those of digital inequality. The odds of leisure activity are greater for singles and people in past relationships than people who are married. Also, the odds of leisure activity are greater for men than for women, a finding that converges with other research, especially in gaming (Schumacher and Morahan-Martin 2001; Willoughby 2008).

Specific Leisure Activities

4.8 The access and usage variables, most of which are positively associated with each leisure activity, are used as controls. This permits the argument: 'controlling for time spent on the Internet and other activities done on the Internet, x has a significant effect on y'.

Table 5. Binary Logistic Regressions for General Internet Leisure Activities

4.9 For each activity there is an intensification of what was seen in the general activity model for leisure. Income is rarely positively associated with any leisure activity. When income is significant, in the case of gaming, we see that with an increase in income there is a decrease in the odds of gaming. In all leisure activities, higher educated groups are less likely than lower educated groups to partake in any given leisure activity. Racial differences are also apparent. These differences are not uniform however. The odds of doing hobbies online increase if a respondent is European-American, while the odds of reading online decrease. One consistent pattern is that Hispanic-Americans and African-Americans tend to share tendencies, such that if one group has greater or lesser odds than European-Americans, so does the other (although in several cases these differences do not reach significance). Other Americans deviate from the minority pattern with respect videos and music.

4.10 Moving to status group variables, we see some structuring of internet by age. However, the cut-off point for age effects appear to be around 50, such that the odds of someone over 50 doing any particular activity is lower than someone between the ages of 18 and 29. This finding was somewhat surprising as it goes against the notion that the Internet as a place for entertainment is the province of the very young. Further, older groups have greater odds of reading than younger groups.

4.11 As with the general usage models, community status is rarely a significant factor, only effecting the watching of videos. Finally, married respondents have consistently lower odds of doing internet activities than others, and males have consistently higher odds of doing internet activities than females.

Discussion

5.1 This research began with the question: What are the predictors of internet leisure patterns? Two perspectives order this research and generated the predictors: digital inequality and status groups. Below I discuss the general understandings gained from this research. Two main points stand out.

5.2 First, both digital inequality and status group perspectives tend to work together to explain all four internet activities. Still, it is clear that leisure activity deviates from other activities, evidenced by the greater influence of status group variables. Even with the greater influence of status groups on leisure, digital inequality predictors are still important. Thus while males and females, young and old, married and single are consuming internet leisure at rates that coalesce with their particular social situation, these activities are not completely divorced from economic, educational and racial/ethnic constraints. I suggest that the best way to understand internet leisure patterns are through a series of profiles that cross-cut digital inequality and status group perspectives. For example, 'Information Seekers'—those with higher odds of using the internet for information—are high income, younger and married. Or, 'Communicators' are highly educated, younger females who are not likely to be African-American.

5.3 Second, internet leisure is best conceptualised as a form of popular culture. Income is not a strong predictor of leisure activities. In all leisure activities, respondents with less than a high school education are more likely than higher educated groups to partake in any given leisure activity. This finding was surprising because of its consistency. When combining the education effects with the lack of income effects, this research suggests that the internet as a medium of entertainment can at least be compared to, and possibly lumped into, the same category as 'popular' culture (Deery 2003), such as blockbuster movies, sitcoms, and pop music (Gans 1999). The label 'popular' is most appropriate here because of the lack of effect of education combined with the notion that media industries are using the Internet to reach people who are consuming their most popular shows or news programs (Deery 2003).

5.4 It goes without saying that internet leisure does not follow the patterns of other forms of internet usage. The age disparity for internet leisure activities begins at later ages, around 50. Indeed respondents between the ages of 30 and 49 often have higher odds of doing an activity than their younger counterparts. Further, African-Americans and Hispanic-Americans, so often lumped together in digital inequality research with the implication that they are disadvantaged vis-à-vis European-Americans, are oftentimes more likely to consume leisure activities than European-Americans.

5.5 The understanding that the lower educated and minority groups are consuming internet leisure at higher rates is cause for optimism and concern. On the one hand, the high rates of leisure activity could be seen as an initial step into more comprehensive usage patterns. Alternatively, the lumping of these groups who have relatively lower cultural and economic capital can lead to the development of symbolic boundaries between a 'low' culture that consumes the products designed for the Internet, and a 'high' culture that distinguish themselves by only employing the Internet for utilitarian purposes.

5.6 Dimaggio et al.(2001) argue that 'Sociology has been slow to take advantage of the unique opportunity to study the emergence of a potentially transformative technology in situ' (329). I have attempted to address this lack of sociological involvement by raising the question: What are the predictors of internet leisure patterns? It appears as if some activities done on the Internet are more a function of cultural differences than socioeconomic barriers. As broadband internet connections continue to rise in the United States, and with broadband users more likely to engage in more diverse types of activities than dial-up users (National Telecommunications and Information Administration 2004), it may be worthwhile to begin an exploration into the stylisation of internet life.

Notes

1I have made a decision to operationalise Digital Inequality using not only income, but also education and race. Education and race are traditionally Weberian status group concepts, especially when presented as a counterargument to Marxian economic predictors of income and class. However, the body of research on internet and technology assumes Digital Inequality and its corresponding predictors of income, education, and race/ethnicity. I am introducing status groups as a relatively newer approach. In this context, the burden of proof falls on Status Groups. Thus I measure Status Groups without education and race/ethnicity.

2The understanding that culture is not passively consumed, but actively engaged with, manipulated, and made to fit in with a social group's daily life is a guiding principal of the Birmingham school of cultural studies. For a review see: Storey (1999).

3One of the categories for race/ethnicity is 'Other Americans'. These were Americans who were not European-American, African-American or Hispanic-American. In essence, this group is composed of Native Americans, Asian-Americans and individuals who chose not to identify their race/ethnicity.

4Occupation would have been a suitable variable to place within digital inequality. However this variable was not measured in the dataset used for this research.

5This weight was supplied in the dataset by the Pew Research center, and is used to compensate for patterns of non-response that could potentially bias results. The weight corrects for this pattern so that the survey sample matches the parameters of the US population.

6There are two possible explanations for this anomaly: (1) respondents are confusing personal entertainment and leisure with another task such as communicating, or (2) the specific tasks given by the survey instrument are not the leisure activities higher educated persons consume.

References

ADAMS, J. and Roscgino, V. J. (2005) 'White Supremacists, Oppositional Culture and the World Wide Web', Social Forces, Vol. 84, No. 2, pp. 759 – 778.

ATTEWELL, P. (2001) 'The First and Second Digital Divides', Sociology of Education, Vol. 74, No. 3, pp. 252 – 259.

BENKLER, Y. (2006) The Wealth of Networks. New Haven, CT: Yale University Press.

BIMBER, B. (2000) 'Measuring the Gender Gap on the Internet', Social Science Quarterly, Vol. 81, No. 3, pp. 868-876.

BOURDIEU, P. (1984) Distinction: A Social Critique of the Judgment of Taste. Cambridge:Harvard University Press.

BRUNSTING, S. and Postmes, T. (2002) 'Social Movement Participation in the Digital Age:Predicting Offline and Online Collective Action', Small Group Research, Vol. 33, No. 5, pp. 525 – 554.

CASTELLS, M. (2001) The Internet Galaxy: Reflections on the Internet, Business, and Society.Oxford: Oxford University Press.

CHAKRABORTY, J. and Bosman, M. M. (2005) 'Measuring the Digital Divide in the United States: Race, Income, and Personal Computers', Professional Geographer, Vol. 57, No. 3, pp. 395 – 410.

CHAN, W. T. and Goldthorpe, J.( 2007) 'Class and Status: the Conceptual Distinction and its Empirical Relevance', American Sociological Review, Vol. 72, No. 4, pp. 512 – 532.

CRANG, M. Crosbie, T. and Graham, S.(2006) 'Variable Geometries of Connection: Urban Digital Divides and the Uses of Information Technology', Urban Studies, Vol. 43, No. 13, pp. 2551 – 2570.

DEERY, J. (2003) 'TV.com: Participatory Viewing on the Web', The Journal of Popular Culture Vol. 37, No. 2., pp., 161 – 183.

DIANI, M. (2000) 'Social Movement Networks Virtual and Real', Information, Communication, and Society, Vol. 3, No. 3., pp. 386 – 401.

DIMAGGIO, P. E., Hargittai, E., Newman, W.R., and Robinson, J. P. (2001) 'Social Implications of the Internet', Annual Review of Sociology, Vol. 27, No. 1, pp. 307 – 336.

DIMAGGIO, P., Hargittai, E.,Celeste, C.,and Shafer, S. (2004) 'Digital Inequality, From Unequal Access to Differentiated Use', in K. Neckerman (editor) Social Inequality, pp. 355 – 400. New York: Russell Sage Foundation.

DURNDELL, A., and Zsolt H. (2002) 'Computer Self Efficacy, Computer Anxiety, Attitudes towards the Internet and Reported Experience with the Internet, by Gender, in an East European Sample', Computers in Human Behavior, Vol. 18, No. 5,pp. 521-535.

GAROFALO, R., Herrick, A., Mustanski, B.S., and Donenberg, G.R. (2007) 'Tip of the Iceberg: Young Men Who Have Sex with Men, the Internet, and HIV Risk', American Journal of Public Health Vol. 97, No. 6, pp. 1113 – 1117.

GATES, H. (2000) 'Black to the Future' Education Week, January 12, P.72

GANS, H. (1999) Popular Culture and High Culture: An Analysis and Evaluation of Taste. New York: Basic Books.

GEFEN, D. and Straub, D.W. (1997) 'Gender Differences in the Perception and use of E-mail: An Extension to the Technology Acceptance Model', MIS Quarterly, Vol. 21, No. 5, pp. 389 – 400.

GERTH, H. H., and Mills, C.W. (1958) From Max Weber: Essays in Sociology. New York: Oxford University Press.

HARGITTAI, E. (2004) 'Internet Access and Use in Context', New Media and Society, Vol. 6, No. 1, pp. 137-143.

HARGITTAI, E. (2008) 'The Role of Expertise in Navigating Links of Influence', in J. Turow and L. Tsui (editors) The Hyperlinked Society. Michigan: The University of Michigan Press

JACKSON, L., Ervin K., Gardner, P., & Schmitt, N. (2001), 'Gender and the Internet: Women Communicating and Men Searching – Statistical Data Included', Sex Roles: A Journal of Research Vol. 44, No. 5/6, pp. 363 – 380.

KATZ, J., E., and Rice, R., E. (2002) Social Consequences of Internet Use: Access, Involvement and Expression. Cambridge: MIT Press

KEEGAN, M. (2004) 'Digital Divide in Computer Access and Use between Poor and Non-Poor Youth', Journal of Sociology & Social Welfare, Vol. 31, No. 2, pp. 91-112.

KOCH, S. C., Meuller, B., Kruse, L., and Zumbach, J. (2005) 'Constructing Gender in Chat Groups',Sex Roles, Vol. 53, No. 1/2, pp. 29 – 41.

KVASNY, Lynette (2006) 'Cultural (Re)production of Digital Inequality in a US Community Technology Initiative', Information, Communication, and Society, Vol. 9, No. 2, pp.160 – 181.

LACEY, M. (2000) 'Clinton Enlists Top-Grade Help for Plan to Increase Computer

Use', New York Times, February 3, p. A25.

LORENCE, D. P. and Park, H. (2006) 'New Technology and Old Habits: The Role of Age as a Technology Chasm' Technology & Health Care Vol. 14, No. 2,: 91-96.

MARTIN, S. P. and Robinson, J. P. (2007) 'The Income Digital Divide: Trends and Predictions for Levels of Internet Use', Social Problems, Vol. 54, No. 1, pp. 1-22.

MORAHAN-MARTIN, J., and Schumacher, P. (2003) 'Loneliness and Social Uses of the Internet', Computers in Human Behavior, Vol. 19, No. 6, pp. 659 – 671.

MORREL, R.W., Mayhorn, C.B., and Bennet, J. (2000) 'A Survey of World Wide Web use in Middle-Aged and Older Adults', Human Factors, Vol. 42: 175 – 182.

MOSSBERGER, K., Tolbert, C.J., Gilbert, M. (2006) 'Race, Place, and Information Technology', Urban Affairs Review, Vol. 41, No. 5, pp. 583 – 620.

NATIONAL Telecommunications and Information Administration. U. S. Department of Commerce. (2000). Falling Through the Net: Toward Digital Inclusion. Washington, DC.

NATIONAL TELECOMMUNICATIONS AND INFORMATION ADMINISTRATION. U. S. DEPARTMENT OF COMMERCE. (2002). A Nation Online: How Americans are Expanding their use of the Internet. Washington, DC.

NATIONAL Telecommunications and Information Administration. U. S. Department of Commerce. (2004). A Nation Online: Entering the Broadband Age. Washington, DC.

ONO, H. and Zavodny, M. (2003) 'Gender and the Internet', Social Science Quarterly, Vol. 84, No. 1, pp. 111-121.

ONO, H. and Zavodny, M. (2008) 'Immigrants, English Ability, and the Digital Divide', Social Science Quarterly, Vol. 36, No. 3, pp. 1135-1155.

RHEINGOLD, H. (1993) The Virtual Community: Homesteading on the Electronic Frontier. Reading, MA: Addison-Wesley Publishing Company.

SCHUMACHER, P. and Morahan-Martin, J. (2001) 'Gender, Internet and Computer Attitudes and Experiences', Computers in Human Behavior, Vol. 17, No. 1, pp. 95-110.

SERVON, L. J., and Nelson, M. K., (2001) 'Community Technology Centers: Narrowing the Digital Divide in Low-Income, Urban Communities', Journal of Urban Affairs, Vol. 23, No. 3/4, pp. 279 – 291.

SPOTTS, T. H., Bowman,M. A., and Mertz, C. (1997) 'Gender and Use of Instructional Technologies', Higher Education, Vol. 34, No. 4, pp. 421 – 426.

STOREY, J. (1999) Cultural Consumption and Everyday Life. New York: Oxford University Press.

VAN EIJCK, K. (2001) ‘Social Differentiation in Musical Taste Patterns’, Social Forces, Vol. 79. No. 3, pp. 1163-1184.

WARSCHAUER, M. (2003) Technology and Social Inclusion: Rethinking the Digital Divide. Cambridge: MIT Press.

WILLOUGHBY, T. (2008) 'A Short-Term Longitudinal Study of Internet and Computer Game Use by Adolescent Boys and Girls: Prevalence, Frequency of Use, and Psychosocial Predictors', Developmental Psychology, Vol. 44, No. 1, pp. 195-204.

WHITTY, M. T., and Mclaughlin, D. (2007) 'Online Recreation: The Relationship between Loneliness, Internet Self-Efficacy and the use of the Internet for Entertainment Purposes'. Computers in Human Behavior, Vol. 23, No. 3, pp. 1435 – 1446.

XIE, B. and Jaeger, P. (2008) 'Older Adults and Political Participation on the Internet: A Cross-cultural Comparison of the USA and China' Journal of Cross-Cultural Gerontology Vol. 23, No. 1, pp. 1-15.